158 Neuroscience Letters, 93 (1988) Elsevier Scientific Publishers Ireland Ltd.

Transcription

1 158 Neuroscience Letters, 93 (1988) Elsevier Scientific Publishers Ireland Ltd. NSL Autoradiographic and electrophysiological evidence for excitatory amino acid transmission in the periaqueductal gray projection to nucleus raphe magnus in the rat Leif Wiklund l, Gila Behzadi 1, Peter Kal+n 2, P. Max Headley 3, Lina S. Nicolopoulos 3, Chris G. Parsons 3 and David C. West 3 ~ Equipe de Neuroanatomie Fonctionnelle, Laboratoire de Physiologie Nerveuse, C.N.R.S., G~f sur- Yvette (France), :Department of Medical Cell Research, University of Lund, Lund (Sweden) and.~department of Physiology, The Medical School, Bristol ( U. K.) (Received 14 April 1988; Revised version received 29 June 1988; Accepted 30 June 1988) Key word~'." Excitatory amino acid transmitter; Nucleus raphe magnus; Periaqueductal gray; Transmitter-selective retrograde labelling; D-Aspartate; Autoradiography; Electrophysiology; Glutamate receptor antagonist; Microelectrophoresis Selective retrograde labelling was used as an autoradiographic method to identify possible excitatory amino acid afferents to nucleus raphe magnus (NRM). Injections of25-50 nl 10-2 or 10 3 M D-[3H]aspar - tate into the NRM resulted in prominent labelling of cells in ventrolateral mesencephalic periaqueductal gray (PAG). Electrophysiologically, stimulation in ventrolateral PAG excited cells in NRM with a latency of 2 12 ms. With many cells, microelectrophoretic application of the excitatory amino acid antagonists, kynurenate and 7-D-glutamyl-glycine, resulted in a reversible reduction of the PAG-evoked response. Selective antagonists of N-methyl-D-aspartate (NMDA) were less effective. It is suggested that neurones in the ventrolateral PAG projecting to NRM utilize an excitatory amino acid or structurally related compound as a transmitter, and that this transmitter acts on receptors of the non-nmda type. The periaqueductal gray (PAG) projection to the nucleus raphe magnus (NRM) is believed to play an important role in descending nociceptive control, opiate and stimulation produced analgesia (for review see ref. 3). Several transmitters have been implicated in the PAG-NRM projection. Beitz [6] demonstrated that some PAG neurones projecting to the NRM were immunoreactive for neurotensin. More recently, it was shown that many PAG cells stain with immunocytochemical markers for excitatory amino acids [9], and behavioural evidence implies excitatory amino Correspondence." L. Wiklund, Equipe de Neuroanatomie Fonctionnelle, Laboratoire de Physiologie Nerveuse, C.N.R.S., Gif-sur-Yvette, France /88/$ Elsevier Scientific Publishers Ireland Ltd.

2 acid transmission of the PAG-NRM projection [1]. Selective retrograde labelling with o-[3h]aspartate [20] has proved a useful method for predicting excitatory amino acid connections [4, 14, 15] (for review see ref. 10). In the present study, we have used this autoradiographic technique to identify possible excitatory amino acid projections from the mesencephalic PAG to NRM, and have subsequently used electrophysiological techniques to test the effects of excitatory amino acid antagonists on neurotransmission in this system. Neuroanatomical experiments. Twelve Sprague-Dawley rats ( g) were used. o-[3h]aspartate (o-[2,3-3h]aspartic acid, Amersham, Ci/mmol) was evaporated under a stream of nitrogen, and redissolved in distilled water or saline to a concentration of 10-2 or 10-3 M. Rats were anaesthetized with pentobarbitone (40 mg/ kg) (n = 5), or a mixture of ketamine (Imalg~ne, Rh6ne M6rieux; 65 mg/kg) and xylazine (Rompun, Bayer; 14 mg/kg) (n = 7), all i.p. Stereotaxic microinjections of 25 or 50 nl o-[3h]asp solution were delivered over 10 min into the NRM through glass micropipettes (outer diameter 20-30/tm) glued to a 0.5/LI Hamilton syringe; pipettes were left in situ for another 5 min before withdrawal. After 6, 12 or 24 h survival, the rats were re-anaesthetized and perfusion fixed; blood was rinsed out with 100 ml Rheomacrodex-NaC1, followed by 1200 ml of 3.5% glutaraldehyde in 0.1 M phosphate buffer, ph 7.4, over 40 min. Dissected brains were kept overnight in the same fixative, and were transferred to 30% sucrose in the same buffer until they sank. Frozen sections were cut at 20 or 30/zm, every fifth section mounted on subbed slides, dried, defatted, and dipped in Kodak NTB-2 emulsion diluted 1:1 with distilled water. Autoradiograms were developed in Dektol after 4 or 8 weeks exposure, stained with Cresyl violet, coverslipped, and evaluated by bright and dark-field microscopy (for details, see ref. 10). Electrophysiological experiments were performed on 15 rats anaesthetized with pentobarbitone (n=4) or pentobarbitone supplemented with ~-chloralose (n= 11). Two Tungsten stimulating electrodes, spaced 2 mm apart and mounted on a micromanipulator, were lowered into the PAG at a ventrocaudal angle of 45 using a dorsal approach; they were positioned initially 1-2 mm anterior and 5 mm dorsal to the interaural line, and were subsequently tracked to obtain the site requiting the lowest stimulation strengths to elicit responses in NRM neurones. Stimulation parameters were /~A and ms at 1 or 2 Hz. Multibarrel micropipettes were inserted into NRM using a ventral approach. The centre barrels of pipettes contained 3.5 M NaC1 and were used for extracellular recording of unitary NRM activity. Side barrels contained combinations of the following drugs for microelectrophoretic administration; kynurenate Na (100 mm in 100 mm NaCI), ~,-o-glutamylglycine Na (DGG; 200 mm), (+)-2-amino-5-phosphonovalerate Na (AP5; 50 mm in 150 mm NaCI), quisqualate Na (5 mm in 195 mm NaCI), kainate (5 mm in 195 mm NaCI) and N- methyl-o-aspartate (NMDA; 100 mm in 100 mm NaCI), all at ph ; ketamine HC1 (50 mm in 150 mm NaC1, ph 5) and Pontamine sky blue (2% in 500 mm sodium acetate, ph 7.7) which was ejected at selected recording sites. In some experiments the effect of intravenous ketamine (4 mg/kg) was tested. Tests were recorded as oscilloscope records, as peristimulus histograms (bin width ms) and as counts of 159

3 160 evoked spikes. Histograms were of 128 or 256 sweeps and were constructed automatically every 2-3 min throughout recording periods. On some cells tests were performed (not illustrated) to check the selectivity, between amino acid analogues, of AP5 and ketamine at the doses tested on the evoked responses. At the end of experiments, the position of stimulating electrodes was marked by electrolytic lesioning. The dissected brains were fixed in 4% formaldehyde, sectioned on a freezing microtome, and positions of stimulating and recording sites identified. The D-[3H]Asp injection sites were usually well centred in the NRM, and were characterized by a central zone of high grain density which progressively decreased towards the periphery (Fig. 1A). In all experiments, retrogradely labelled cells appeared in the mesencephalic PAG (Fig. I B, C). The number of labelled PAG cells and their labelling intensity depended on the amount of D-[3H]Asp injected into NRM; in cases injected with 10-2 M solution several hundred labelled cell profiles were detected in each series of autoradiograms. The labelled PAG cells were of small to moderate size, and often demonstrated labelling of a few primary dendritic processes (Fig. 1C). The labelled cells were concentrated in the ventrolateral portion of Fig. I. Autoradiographic labelling observed after injection of 50 nl D-[3H]Asp, 10 2 M, into the nucleu~ raphe magnus. A: low-power bright-field view of the injection site (arrow) at the level of the facial nucleus. P, pyramid; asterisk, cut in tissue for identification of left hand side. Bar= 1 mm. B: low-power dark-field view of retrogradely labelled cells (arrows) in the ventrolateral periaqueductal gray. Aq, cerebral aqueduct. Bar = 200/~m. C: high-power bright-field micrograph showing a well labelled cell in PAG (arrow), and several neighbouring unlabelled cells (arrowheads). Bar = 50/xm.

4 161 PAG. Comparison with the careful mapping of serotonergic cells by Descarries et al. [I 2], indicated that a few D-[3H]Asp-labelled cells were situated within the more medial region of PAG that contain serotonergic cells. In contrast, very few labelled cells were found in the dorsomedial part of PAG, which is also known to project to the NRM [8]. Other mesencephalic regions known to project to the NRM, such as the cuneiform nucleus and superior colliculus [8], showed no or very few labelled cells. In electrophysiological experiments, cells in the NRM which responded to PAG stimulation were found readily. Most of the marked stimulation sites were in the ventrolateral PAG, corresponding to the region where an accumulation of labelled cells had been observed in the D-[3H]Asp tracing experiments. The latency of evoked responses varied between cells from 2 to 12 ms. Forty-five NRM cells responding to PAG stimulation were tested with microelectrophoretic administrations of antagonists of varying selectivity for the 3 commonly recognized receptors for excitatory amino acids [2, 21]. Fig. 2 illustrates the effect of kynurenate, an excitatory amino acid antagonist which does not distinguish between different subclasses of glutamate receptors [19], on the PAG-evoked responses of an NRM neurone. Kynurenate reduced the shorter latency component of PAG responses, while the later responses were less affected. Another non-selective excitatory amino acid antagonist, DGG, also decreased PAG-evoked responses but higher electrophoretic currents were usually required than with kynurenate. In all, about one quarter of the cells tested with kynurenate ( na) demonstrated a marked reduction (~> 50%) of the PAG-evoked responses, another quarter of the cells showed a less pronounced effect, while half of the cells showed no or very little effect CONTROL ~ KYN 100nA : "~.. ~ OFF 1 min PAG A1.7 NRM P2.3 0 t' 5 msec Fig. 2. Responses of an NRM neurone to stimulation of the PAG. NRM neurone location 2.3 mm posterior to the interaural line, as indicated by the insert diagram. PAG electrode sites shown on insert diagram; electrical stimuli were of 160/tA and 0.5 ms repeated at 2 Hz. The evoked responses were averaged in a peristimulus time histogram (0.1 ms bins) collected over 256 consecutive sweeps. The arrows under the histograms indicate the onset of the stimulus. The record is suppressed during the period of the stimulus and resulting artefact. After the control period (left histogram) kynurenate was ejected microelectrophoretically at 100 na for I min before the start of the next accumulation (centre histogram). One min after the end of the kynurenate ejection the right hand histogram collection was started. Note that the shorter latency part of the evoked response was reduced preferentially during the kynurenate ejection.

5 162 of the antagonist. The responses of most NRM cells were resistant to the selective NMDA antagonists ketamine or AP5. The selectivity of retrograde labelling with D-[3H]Asp depends on how selective the axon terminal uptake mechanisms are for neurones which utilize excitatory amino acid (or related) neurotransmitters [10]. Since this uptake system fails to distinguish between L-Glu, L-Asp and n-asp [11], the method does not indicate whether glutamate or aspartate is more likely to be the endogenous transmitter. Moreover, it has been claimed that a dipeptide active at excitatory amino acid receptors, N-acetylaspartyl-glutamate, may be the transmitter of certain neuronal pathways. Recently, it has been suggested that such neurones may demonstrate glutamate uptake [7]. It is, therefore, possible that neurones utilizing this dipeptide could be labelled retrogradely with D-[3H]Asp. The characteristics of the PAG-evoked responses recorded electrophysiologically in NRM were similar to those reported previously [5, 16-18]. The spread of latencies of PAG-evoked responses is consistent with the conduction velocities of PAG-NRM axons in rats which vary between 0.4 and 9.6 m/s [18] so that stimulation in the PAG would evoke monosynaptic potentials with a latency range as wide as ms; monosynaptically evoked spikes are therefore likely to have a variable latency. Our autoradiographic experiments with D-[3H]Asp indicate that there is a direct excitatory amino acid projection from PAG to NRM, but electron-microscopical investigation of identified PAG terminals would be necessary to demonstrate if they synapse directly on the NRM cells and their dendrites. The preferential sensitivity of the earlier components of evoked responses to the amino acid antagonists is however supportive of amino acid mediation of the PAG projection neurones. The susceptibility of responses to kynurenate and DGG provides no evidence on the amino acid receptor types mediating the response. Selective NMDA antagonists were, however, markedly less effective, whether administered electrophoretically (AP5 and ketamine) or intravenously (ketamine) at NMDA-blocking doses [2]. These observations therefore suggest that the excitatory amino acid (or structurally related) neurotransmitter acts on receptors of the non-nmda type. Interestingly, the region of the PAG which was preferentially labelled in our autoradiographic experiments, and stimulated in our electrophysiological tests, seems to correspond to the 'pure analgesic region' in the ventrolateral PAG described by Fardin et al. [13] in their topographic analysis of stimulation produced analgesia. Moreover, Aimone and Gebhart [1] have demonstrated in behavioural experiments that the antinociceptive effects of stimulating the ventrolateral PAG can be reduced by microinjections of excitatory amino acid antagonists into the NRM. These combined observations indicate that the component of the PAG-NRM projection identified by the present study plays a significant role in descending nociceptive control. We thank Prof. M. Cu6nod for his contribution to the early part of this study, Mrs. G. Ghilini for skilled technical assistance, and C.N.R.S., La Fondation pour la Recherche M6dicale, the Swedish MRC (Grant 06535), the English MRC and the Wellcome Trust for financial support.

6 1 Aimone, L.D. and Gebhart, G.F., Stimulation-produced spinal inhibition from the midbrain in the rat is mediated by an excitatory amino acid neurotransmitter in the medial medulla, J. Neurosci., 6 (1986) Anis, N.A., Berry, S.C., Burton, N.R. and Lodge, D., The dissociative anaesthetics, ketamine and phencyclidine, selectively reduce excitation of central mammalian neurones by N-methyl-aspartate, Br. J. Pharmacol., 79 (1983) Basbaum, A.I. and Fields, H.L., Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry, Annu. Rev. Neurosci., 7 (1984) Beart, P.M., Nicolopoulos, L.S., West, D.C. and Headley, P.M., An excitatory amino acid projection from ventromedial hypothalamus to periaqueductal gray in the rat: autoradiographic and electrophysiological evidence, Neurosci. Lett., 85 (1988) Behbehani, M.M. and Fields, H.L., Evidence that an excitatory connection between the periaqueductal gray and nucleus raphe magnus mediates stimulation produced analgesia, Brain Res., 170 (1979) Beitz, A.J., The sites of origin of brain stem neurotensin and serotonin projections to the rodent nucleus raphe magnus, J. Neurosci., 2 (1982) Blakely, R.D., Ory-Lavoll6e, L., Thompson, R.C. and Coyle, J.T., Synaptosomal transport of radiolabel from N-acetyl-aspartyl-[3H]glutamate suggests a mechanism of inactivation of an excitatory neuropeptide, J. Neurochem., 47 (1986) Carlton, S.M., Leichnetz, G.R., Young, E.G. and Mayer, D.J., Supramedullary afferents of the nucleus raphe magnus in the rat: a study using the transcannula HRP gel and autoradiographic techniques, J. Comp. Neurol., 214 (1983) Clements, J.R., Madl, J.E., Johnson, R.L., Larson, A.A. and Beitz, A.J., Localization of glutamate, glutaminase, aspartate and aspartate aminotransferase in the rat midbrain periaqueductal gray, Exp. Brain Res., 67 (1987) Cu6nod, M. and Streit, P., Neuronal tracing using retrograde migration of labeled transmitter-related compounds. In A. Bj6rklund and T. H6kfelt (Eds.), Handbook of Chemical Neuroanatomy, Vol. 1, Elsevier, Amsterdam, 1983, pp Davies, L.P. and Johnston, G.A.R., Uptake and release of D- and L-aspartate by rat brain slices, J. Neurochem., 26 (1976) Descarries, L., Watkins, K.C., Garcia, S. and Beaudet, A., The serotonin neurons in nucleus raphe dorsalis of adult rat: a light and electron microscope radioautographic study, J. Comp. Neurol., 207 (1982) Fardin, V., Oliveras, J.L. and Besson, J.M., A reinvestigation of the analgesic effects induced by stimulation of the periaqueductal gray matter in the rat. I. The production of behavioural side effects together with analgesia, Brain Res., 306 (1984) Kal6n, P., Karlson, M. and Wiklund, L., Possible excitatory amino acid afferents to nucleus raphe dorsalis of the rat investigated with retrograde wheat germ agglutinin and D-[3H]aspartate tracing, Brain Res., 360 (1985) Kal6n, P., Pritzel, M., Nieoullon, A. and Wiklund, L., Further evidence for excitatory amino acid transmission in the lateral habenular projection to the rostral raphe nuclei: lesion-induced decrease of high affinity glutamate uptake, Neurosci. Lett., 68 (1986) Lovick, T.A., West, D.C. and Wolstencroft, J.H., Responses of raphespinal and other bulbar raphe neurones to stimulation of the periaqueductal gray in the cat, Neurosci. Lett., 8 (I 978) Maciewicz, R., Sandrew, B.B., Phipps, B.S., Poletti, C.E. and Foote, W.E., Pontomedullary raphe neurons: intracellular responses to central and peripheral stimulation, Brain Res+, 293 (1984) Shah, Y. and Dostrovsky, J.O., Electrophysiological evidence for a projection of the periaqueductal gray matter to nucleus raphe magnus in cat and rat, Brain Res., 193 (1980) Stone, T.W. and Connick, J.H., Quinolinic acid and other kynurenines in the central nervous system, Neuroscience, 15 (1985) 597~ Streit, P., Selective retrograde labeling indicating the transmitter of neuronal pathways, J. Comp. Neurol., 191 (1980) Watkins, J.C. and Evans, R.H., Excitatory amino acid transmitters, Annu. Rev. Pharmacol. Toxicol., 21 (1981)